Project Summary Integrated Approaches to close the loop in Type 1 diabetes. The RFA 08-012 emphasizes closing the loop in type 1 diabetes mellitus (T1DM), an endeavor that will depend on developing physiological models that will evaluate the reasons for glycemic variability using state of the art technologic approaches so that continuous subcutaneous insulin infusion (CSII) algorithms are better informed. The physiology studies need to be directed towards understanding the impact of variability introduced by circadian and intra-individual differences in post prandial insulin action, variability in the timing and rate of appearance of ingested carbohydrates into the systemic circulation and variability introduced by changes in physical activity. Using innovative specific activity clamps to minimize non steady state errors, we have recently developed and validated a triple tracer technique to measure post prandial peripheral and hepatic insulin action and rate of appearance of meal derived glucose. We have also developed and validated tools (eg., accelerometers, physical activity monitoring systems) to accurately capture daily physical activity. In this context, and utilizing the above techniques, the following specific aims will be addressed: 1) We will determine whether circadian pattern of post prandial insulin action and meal glucose appearance occurs in T1DM and the extent to which it differs from nondiabetic healthy subjects. We will also take this opportunity to further develop and validate a novel single tracer method to measure post prandial glucose kinetics. 2) We will explore if gastric emptying rate can predict post prandial meal glucose appearance, insulin action and 24 hour glucose variability in T1DM. To do so, we will screen for abnormalities in gastric emptying then use the tripe tracer method (or single tracer model if validated in aim 1) to determine if asymptomatic changes in gastric emptying are accompanied by changes in the timing and systemic rate appearance of ingested glucose and if so, the extent to which changes in the pattern of meal appearance predict post prandial hyperglycemia and glucose variability as measured using CGM parameters. This is an important variable to consider since abnormal gastric motility is frequently observed in asymptomatic individuals with T1DM. 3) We will evaluate the impact of low and moderate intensity physical activity on glucose variability, post prandial meal glucose appearance and insulin action in T1DM. To do so, we will assess the impact of low and moderate intensity exercise (measured by accelerometers) on glucose variability using CGM parameters (high and low blood glucose indices), meal glucose appearance and post prandial insulin action. We will take this opportunity to further modify tracer technique if necessary to clamp specific activity in order to accurately estimate insulin action in the post prandial state. Summary and Significance: Glucose variability and frequent hypoglycemia are major factors that limit optimal subcutaneous insulin delivery in T1DM. In this application, we propose to utilize and further innovate existing cutting edge techniques to develop a physiological model whereby daily variations in post prandial insulin action, gastric motility and physical activity captured with precise physical activity monitoring systems, e.g., accelerometers, are integrated to facilitate development of a model predictive control algorithm of individualized closed loop system of insulin delivery.